WikiLeaks Document Release http://wikileaks.org/wiki/CRS-RS22124 February 2, 2009 Congressional Research Service Report RS22124 Potential House Apportionment Following the 2010 Census Based on Census Bureau Population Projections David C. Huckabee, Government and Finance Division April 26, 2005 Abstract. The Census Bureau's 2005 population projections for the year 2010 raise the possibility of potentially significant changes in the allocation of Representatives among the states. If the projections for the year 2010 presage the actual Census, 10 seats will shift, affecting a total of 15 states. CRS experience with prior Census Bureau population projections suggests, however, they are an imperfect predictor of actual number of Representatives states will be granted after a census. Order Code RS22124 April 26, 2005 CRS Report for Congress Received through the CRS Web Potential House Apportionment Following the 2010 Census Based on Census Bureau Population Projections1 Royce Crocker Specialist in American National Government Government and Finance Division http://wikileaks.org/wiki/CRS-RS22124 Summary The Census Bureau's 2005 population projections for the year 2010 raise the possibility of potentially significant changes in the allocation of Representatives among the states. If the projections for the year 2010 presage the actual Census, 10 seats will shift, affecting a total of 15 states. CRS experience with prior Census Bureau population projections suggests, however, they are an imperfect predictor of actual number of Representatives states will be granted after a census. Although the Bureau of the Census estimates the population for each state annually, state-level population projections are usually issued once each decade. On April 21, 2005, the Bureau released projected state populations for the year 2010. If the House of Representatives were to be reapportioned based on these projected numbers, 10 seats would shift among 15 states from the official apportionment following the 2000 Census. Arizona, California, Georgia, Nevada and Utah would each gain one seat; Florida would gain two seats; and Texas would gain three seats. The following states would lose one seat: Alabama, Illinois, Iowa, Massachusetts, Missouri, and Pennsylvania; New York and Ohio would lose two seats if these projections prove to be accurate. Caveats The official apportionment based on the 2010 Census will probably differ from the trial apportionment based on the projections. This is because population projections are of uncertain accuracy. First, projections for large geographic units are more likely to be accurate than those for smaller units. Thus, a projection of the total U.S. population is likely to be more accurate than one for an individual state. Also, adding or subtracting a small number of 1 This report originally was authored by David C. Huckabee, who has retired from CRS. Congressional Research Service ~ The Library of Congress CRS-2 people from a state's population can make a difference in whether or not a seat is assigned to that state. Second, the assumptions that underlie the projections and that can have a significant effect on the resulting apportionment estimates may prove to be erroneous. For example, the 1983 population projection for California for 1990 was 27,525,600, compared to 29,839,250 in the actual 1990 census count -- the 1983 projection suggested a four-seat gain, but the 1990 numbers resulted in a seven-seat increase of Representatives for California. Furthermore, the Census Bureau population projections for 2000 were imperfect predictors of the actual reapportionment of the House of Representatives. Ten states had different totals of Representatives after the census than had been expected based on the population projections.2 Third, population projections are not directly comparable to Census figures. For example, following the practice of the annual population estimates, projections are computed for July 1 of the projection year, whereas the Census is taken on April 1 of each year ending in zero. Thus, the date used for the projection is three months later than the 2000 Census. http://wikileaks.org/wiki/CRS-RS22124 Fourth, a further complicating factor in using projections to compute prospective apportionments concerns the status of federal employees who are stationed abroad (chiefly military personnel and their dependents, totaling 574,330 persons in 2000). In 2000, the Census Bureau included these people in the populations used to reapportion the House.3 The projections used to calculate the prospective apportionment in this report are not adjusted to account for federal employees stationed abroad. Nevertheless, as imperfect as population projections are, they provide a rough basis for estimating what representation in the House may be after the next reapportionment. Tables Table 1 sets out the state populations used to reapportion the House of Representatives after the 2000 Census (the April 1, 2000, census apportionment population), and the Census Bureau's July 1, 2010 population projection. It also illustrates the change from 2000 (shown by total and percent), the current House seat allocation, and what it would be if the House were to be reapportioned based on these population projections. 2 U.S. Bureau of the Census, Population Projections for States, by Age, Sex, Race, and Hispanic Origin: 1993 to 2020, Current Population Reports, P25-1111, (Washington: 1994). The projections suggested that Arizona would have received seven seats instead of eight it actually received. Other states would have changed as follows if the projections had been used to apportion seats rather than the actual census results: California would have received 55 seats, rather than 53; Colorado 6, rather than 7; Florida 24, rather than 25; Georgia 12, rather than 13; Indiana 10, rather than 9; Massachusetts 9, rather than 10; Montana 2, rather than 1; North Carolina 12, rather than 13; and Washington 10, rather than 9 seats. 3 See CRS Report RS20768, House Apportionment 2000: States Gaining, Losing, and on the Margin. by David C. Huckabee, p. 3. This report describes how the equal proportions formula allocates Representatives among the states. CRS-3 Table 1. Possible Apportionment of Seats in the House of Representatives Based on the 2000 Census and 2010 Census Bureau Population Projections 2000 Census 2010 population projection Expected change Apportionment No. of Projected from 2000 No. of Seat change State populationa Seats population Total Percent % seats from 2000b AL 4,461,130 7 4,596,330 135,200 3.03 6 -1 AK 628,933 1 694,109 65,176 10.36 1 AZ 5,140,683 8 6,637,381 1,496,698 29.11 9 +1 AR 2,679,733 4 2,875,039 195,306 7.29 4 CA 33,930,798 53 38,067,134 4,136,336 12.19 54 +1 CO 4,311,882 7 4,831,554 519,672 12.05 7 CT 3,409,535 5 3,577,490 167,955 4.93 5 DE 785,068 1 884,342 99,274 12.65 1 FL 16,028,890 25 19,251,691 3,222,801 20.11 27 +2 GA 8,206,975 13 9,589,080 1,382,105 16.84 14 +1 http://wikileaks.org/wiki/CRS-RS22124 HI 1,216,642 2 1,340,674 124,032 10.19 2 ID 1,297,274 2 1,517,291 220,017 16.96 2 IL 12,439,042 19 12,916,894 477,852 3.84 18 -1 IN 6,090,782 9 6,392,139 301,357 4.95 9 IA 2,931,923 5 3,009,907 77,984 2.66 4 -1 KS 2,693,824 4 2,805,470 111,646 4.14 4 KY 4,049,431 6 4,265,117 215,686 5.33 6 LA 4,480,271 7 4,612,679 132,408 2.96 7 ME 1,277,731 2 1,357,134 79,403 6.21 2 MD 5,307,886 8 5,904,970 597,084 11.25 8 MA 6,355,568 10 6,649,441 293,873 4.62 9 -1 MI 9,955,829 15 10,428,683 472,854 4.75 15 MN 4,925,670 8 5,420,636 494,966 10.05 8 MS 2,852,927 4 2,971,078 118,151 4.14 4 MO 5,606,260 9 5,922,078 315,818 5.63 8 -1 MT 905,316 1 968,598 63,282 6.99 1 NE 1,715,369 3 1,768,997 53,628 3.13 3 NV 2,002,032 3 2,690,531 688,499 34.39 4 +1 NH 1,238,415 2 1,385,560 147,145 11.88 2 NJ 8,424,354 13 9,018,231 593,877 7.05 13 NM 1,823,821 3 1,980,225 156,404 8.58 3 NY 19,004,973 29 19,443,672 438,699 2.31 27 -2 NC 8,067,673 13 9,345,823 1,278,150 15.84 13 ND 643,756 1 636,623 -7,133 -1.11 1 OH 11,374,540 18 11,576,181 201,641 1.77 16 -2 OK 3,458,819 5 3,591,516 132,697 3.84 5 OR 3,428,543 5 3,790,996 362,453 10.57 5 PA 12,300,670 19 12,584,487 283,817 2.31 18 -1 RI 1,049,662 2 1,116,652 66,990 6.38 2 SC 4,025,061 6 4,446,704 421,643 10.48 6 SD 756,874 1 786,399 29,525 3.90 1 TN 5,700,037 9 6,230,852 530,815 9.31 9 TX 20,903,994 32 24,648,888 3,744,894 17.91 35 +3 UT 2,236,714 3 2,595,013 358,299 16.02 4 +1 VT 609,890 1 652,512 42,622 6.99 1 CRS-4 2000 Census 2010 population projection Expected change Apportionment No. of Projected from 2000 No. of Seat change State populationa Seats population Total Percent % seats from 2000b VA 7,100,702 11 8,010,245 909,543 12.81 11 WA 5,908,684 9 6,541,963 633,279 10.72 9 WV 1,813,077 3 1,829,141 16,064 0.89 3 WI 5,371,210 8 5,727,426 356,216 6.63 8 WY 495,304 1 519,866 24,562 4.96 1 281,424,177 435 308,405,442 26,981,265 9.59 435 Notes: a. U.S. Bureau of the Census, Population Projections Branch, Population Division, Florida, California and Texas to Dominate Future Population Growth, Census Bureau Reports, Census Bureau Press Release CB05-52, April 21, 2005. (Seat allocations computed by CRS .) b. Numbers following + and - signs represent net gain or loss in projected seats over 2000 levels. Priority Lists and Seat Assignments http://wikileaks.org/wiki/CRS-RS22124 The reapportionment process for the House relies on rounding principles, but the actual procedure involves computing a "priority list" of seat assignments for the states. The Constitution allocates the first 50 seats because each state must have at least one Representative. A priority list assigns the remaining 385 seats for a total of 435. Table 2 displays the end of the "priority list" that would be used to allocate Representatives based on 2010 projections. The law only provides for 435 seats in the House, but the table illustrates not only the last seats assigned by the apportionment formula (ending at 435), but the states that would just miss getting additional representation.4 Table 2. Population Needed to Gain or Lose a Seat Using Census Bureau Population Projections for 2010 2000 Priority apportionment Population needed to rank State Seat population Priority valuea gain or lose seatb 420 CA 52 38,067,134 739,202.40 -1,543,551 421 IL 18 12,916,894 738,409.65 -510,451 422 TX 34 24,648,888 735,869.63 -892,358 423 TN 9 6,230,852 734,312.62 -212,841 424 NY 27 19,443,672 733,854.07 -652,445 425 FL 27 19,251,691 726,608.22 -460,464 426 CA 53 38,067,134 725,121.12 -834,292 427 MN 8 5,420,636 724,362.69 -113,249 428 NE 3 1,768,997 722,189.92 -31,747 429 NJ 13 9,018,231 722,036.17 -159,960 430 MI 15 10,428,683 719,646.94 -150,969 431 PA 18 12,584,487 719,407.20 -178,044 432 TX 35 24,648,888 714,535.55 -183,053 4 Please note that the figures in table 2 for the "population needed to gain or lose a seat" are misleading because it is unlikely that one state's population total would be adjusted without others changing as well. Since the method of equal proportions used to allocate seats in the House uses all state populations simultaneously, changes in several state populations may also result in changes to the "population needed to gain or lose a seat." CRS-5 2000 Priority apportionment Population needed to rank State Seat population Priority valuea gain or lose seatb 433 LA 7 4,612,679 711,751.64 -16,348 434 CA 54 38,067,134 711,566.24 -125,030 435 GA 14 9,589,080 710,789.33 -21,048 Last seat assigned by law 436 AL 7 4,596,330 709,229.12 +10,112 437 NY 28 19,443,672 707,159.50 +99,804 438 OH 17 11,576,181 701,909.06 +146,457 439 MA 10 6,649,441 700,912.28 +93,702 440 FL 28 19,251,691 700,177.22 +291,785 441 AZ 10 6,637,381 699,641.05 +105,762 442 CA 55 38,067,134 698,508.84 +669,259 443 IL 19 12,916,894 698,465.63 +227,905 444 MO 9 5,922,078 697,923.27 +109,172 445 VA 12 8,010,245 697,202.23 +156,104 446 MD 9 5,904,970 695,907.08 +126,280 447 TX 36 24,648,888 694,403.72 +581,631 448 NC 14 9,345,823 692,757.94 +243,257 http://wikileaks.org/wiki/CRS-RS22124 449 OR 6 3,790,996 692,137.88 +102,158 450 WA 10 6,541,963 689,583.11 +201,180 Notes: a. Each state's claim to representation in the House is based on a "priority value" determined by the following formula: PV = P / [n( n - 1 )]½; where PV = the state's priority value, P = the state's population, and n = the state's nth seat in the House. For example, the priority value of Alabama's 7th seat is: PVAL7 = 4,596,330 / [ 7(7 - 1 ) ]½ = 4,596,330 / [ 42 ]½ = 4,596,330 / 6.480741 = 709,229.12 The actual seat assignments are made by ranking all of the states' priority values from highest to lowest until 435 seats are allocated. b. These figures represent the population a state would either need to lose in order to drop below the 435th seat cutoff, or to gain to rise above the cutoff. If, in the case of Alabama, the population projection had yielded 10,111 more persons, the state's priority value would increase to 710,789.43 which would result in a new sequence number of 435 because Georgia's 14th seat would now occupy the 436th position in the priority list. Source: Computations by CRS using Census Bureau 2010 population estimates. See CRS Report RL30711, The House Apportionment Formula in Theory and Practice, by David C. Huckabee, for an explanation of the formula for allocating House seats. Options for States Losing Seats Apportionment counts transmitted by the Census Bureau to the President after a decennial census (who then sends them to Congress) are considered final. Thus, most states that will lose seats after the 2010 Census will have only one possible option for retaining them: urge Congress to increase the size of the House. Any other option, such as changing the formula used in the computations or changing the components of the CRS-6 apportionment population (such as omitting the foreign-based military and federal civilian employees) will only affect a small number of states if the House stays at 435 seats.5 The 435 seat limit was imposed in 1929 by 46 Stat. 21, 26-27. Altering the size of the House would require a new law setting a different limit. Article 1, Section 2 of the Constitution establishes a minimum House size (one Representative for each state), and a maximum House size (one for every 30,000, or 9,380 based on the 2000 Census). An increase of the size of the House to 473 would have resulted in no states losing seats they held from the 103rd to the 107th Congresses. Those states retaining seats through an increase in the House size would not have been able to retain their pre-2000 Census proportional share of House seats, because other states would also have their delegations become larger. At a House size of 473, California's delegation size, for example, would have been 57 instead of 53 seats.6 http://wikileaks.org/wiki/CRS-RS22124 5 After the 1990 Census, Montana and Massachusetts challenged the apportionment formula and the inclusion of the foreign-based military and civilians in the apportionment population. The Supreme Court affirmed the constitutionality of the equal proportions formula and the inclusion of the foreign-based military and civilians in the counts in two separate cases: U.S. Dept. of Commerce v. Montana 112 S.Ct. 1415 (1992) and Franklin v. Massachusetts 112 S.Ct. 2767 (1992). 6 For a fuller discussion of this topic see U.S. Library of Congress, Congressional Research Service, The House Apportionment Formula in Theory and Practice, by David C. Huckabee, CRS Report RL30711 (Washington: July 11, 1995).